The Drosophila larva is composed of a series of discrete head, thoracic and abdominal segments. The Antannapedia locus (Antp) of the Antannapedia Complex (ANT-C) is required for proper differentiation of the larval thoracic segments. Genes of the bithorax complex (BX-C) are required for the morphogenesis of the metathorax and abdominal segments. The embryonic segments and segment anlage that accumulate transcripts specified by ANT-C and BX-C correspond to those that are most affected by mutations of these genes. Restricted spatial expression of ANT-C and BX-C appears to involve hierarchial interactions of the genes contained within these complexes. BX-C- embryos display an epidermal and neural transformation of the metathorax and first seven abdominal segments into the homologous tissues of the mesothorax. It appears that this transformation results from a posterior extension of the normally thoracic realm of Antp+ expression since transcripts specified by Antp accumulate in the metathoracic and first seven abdominal ganglia of BX-C- embryos. The experiments described in this proposal will permit a detailed identification of the BX-C genes involved in the inhibition of Antp expression in the posterior ganglia of wild-type embryos. In addition, we will use P-mediated transformation to define the sequences within the Antp transcriptional unit that are required for interaction with BX-C. Additional homeotic loci appear to be involved in segmental determination of the anterior embryonic regions that will form the prothorax and head. By localizing transcripts specified by the Scr locus and putative 'head forming' loci within mutant embryos that are deficient for various ANT-C and BX-C genes we will determine whether the anterior realms of expression that are thought to be acquired by these genes result from inhibitory interactions with Antp and bithorax products. Garcia-Bellido has proposed that homeotic genes control developmental pathways by the activation of 'realisator loci' which in turn specify morphogenetic cell properties such as surface recognition characteristics. A prediction of this model is that particular genes become selectively expressed in the mesothoracic ganglion of developing embryos in response to Antp+ products. The experiments proposed in this grant should allow us to test this model by isolating and characterizing neural-specific genes that are present in BX-C embryos and absent in Antp- embryos.